Make 64-bit clean

(matt)

Use the right file.

(matt)

Use uintptr_t instead of int.

(matt)

Use more appropriate type.

(matt)

Make this 64-bit aware.

(matt)

Add pmap_ste_spill

(matt)

Make this work.

(matt)

Add trapframe32 for COMPAT_NETBSD32

(matt)

Add ALIGNBYTES32 for COMPAT_NETBSD32

(matt)

Use uintptr_t for bus_size_t, bus_addr_t, bus_space_handler_t since uint32_t
doesn't really work too 64-bit platforms.

(matt)

support PPC_OEA64

(matt)

Add mcontext32_t

(matt)

Add cmp{ptr,long,reg}{,l}{,i}

(matt)

GC __raise.

(joerg)

GC writeback_dcache_line and clear_dcache_line.

(joerg)

Allow CTLTYPE_INT and CTLTYPE_QUAD to be read and written as either 4 or 8
  byte values regardless of the type.
64bit writes to 32bit variables must be valid (signed) values.
32bit reads of large values return -1.
Amongst other things this should fix libm's code that reads machdep.sse
  as a 32bit int, but I'd changed it to 64bit (to common up some code).

(dsl)

remove the GCC 4 EXTERNAL_GCC_SUBDIR, and switch GCC 4.8 to use gcc.old.

(mrg)

define GCCDIST properly for GCC 4.1, and since it is now the same
as GCCSRCDIR remove the latter.

(mrg)

Global variable is never a null pointer.

(joerg)

Cast argument to abs, not the result.

(joerg)

offs_from_utc is known to be small, so explicitly cast it to int.

(joerg)

Use ## for concatenation

(matt)

Use abs for difference of integers.

(joerg)

Use labs for long arguments and fabs for double ones.

(joerg)

Use _ENTRY

(matt)

time_t is larger than int, so don't use abs.

(joerg)

Member will never be null, so remove check.

(joerg)

Remove tautological check.

(joerg)

Cast strlen of constant down to avoid argument widening. Change an abs
to labs based on the arguments.

(joerg)

Provide labs, llabs and imaxabs for kernel use.

(joerg)

Use the CALL() macro.

(matt)

Don't use abs on unsigned values.

(joerg)

Expect long long to be larger than uint32_t and use llabs after casting
the uint32_t.

(joerg)

Timestamps can be quite large, so use llabs.

(joerg)

Avoid promotion in subexpressions.

(joerg)

Don't use integer downcasts, use correctly sized constants for each test.

(joerg)

The current implementation of vn_lock() is racy.  Modification of
the vnode operations vector for active vnodes is unsafe because it
is not known whether deadfs or the original file system will be
called.

- Pass down LK_RETRY to the lock operation (hint for deadfs only).

- Change deadfs lock operation to return ENOENT if LK_RETRY is unset.

- Change all other lock operations to check for dead vnode once
  the vnode is locked and unlock and return ENOENT in this case.

With these changes in place vnode lock operations will never succeed
after vclean() has marked the vnode as VI_XLOCK and before vclean()
has changed the operations vector.

Adresses PR kern/37706 (Forced unmount of file systems is unsafe)

Discussed on tech-kern.

Welcome to 6.99.33

(hannken)

Add powerpc64 support

(matt)

Fix typo in comment BEAGLEBOARD->CUBIEBOARD

(reinoud)

Add P2SZREG to be usign with .p2align

(matt)

Use gcc,compat, directory creation only depends on the MKCOMPAT switch.

(joerg)

Currently dead vnodes still reside on the vnodelist of the file system
they have been removed from.

Create a "dead mount" that takes dead vnodes until they get freed.

Discussed on tech-kern.

(hannken)

For now, provide a path to the gcc 4.1 distribution as well

(martin)

We have to ensure the string is NUL-terminated and of the expected
length to avoid copying uninitialized data.

ok christos@

(maxv)

Add atomic_cas_64 support for ARM EABI on V5TE and V5TEJ cpus.
(strd is atomic).

(matt)

Add a method to test the compiler for things like LDREX availability,
LDRD availability, THUMB2, EABI.

(matt)

Increase the image size to 1600KB.

(isaki)

fix some paths.

(mrg)

Explicitly remove .note.netbsd.mcmodel sections (eroneously generated
on sparc64 during intermediate linking)

(martin)

ipsec-tools-0.8.2

(tteras)

From Adam Majer <adamm@zombino.com>: Support IPv6 in X509 subjectAltName

(tteras)

fix another dated DIST.

(mrg)

fix typo.

(hkenken)

G/c old unused sh3 csu.  Requested by joerg@

(uwe)

Checking the return value of an allocator works better, when looking at
the stored pointer.

(joerg)

abs on unsigned values is a nop.

(joerg)

Remove impossible checks.

(joerg)

A member of a non-null struct pointer can't be null.

(joerg)

AF_MAX isn't an address family so don't include it in the generated table.

(ginsbach)

To build elftosb with clang++, it needs more than nl_types.h provides.
XXX to workaround with, if compiling on __NetBSD__ just include the real
nl_types.h

(matt)

Use an abs wide enough to cover the types involved.

(joerg)

struct member can't be null, if the pointer itself is valid.

(joerg)

Retire sources from old build location.

(joerg)

Update compiler-rt glue for the new source layout.

(joerg)

delete 'difz' as well, in case i've left it.

XXX: consider running 'cvs up -dPA -I\! -I.CVS -ICVS' first?

(mrg)

oops, avoid re-setting DIST to something that might be wrong now.

(mrg)

oops, did not mean to commit this.

(mrg)

support 601, from scole_mail

(macallan)

fix a typo (from cegger)
fix the date call (from tn)

(mrg)

Remove hardcoded VM_MIN_ADDRESS and instead query current value via
sysctl vm.minaddress.

(martin)

Fix copy & pasto

(martin)

Only track kenter_pa'ed pages if arm_cache_prefer_mask is non-zero.

(matt)

Initialize mvsoctmr_freq earlier (in mvsoctmr_attach) to avoid division
by zero in calculation of the watchdog parameters.

(martin)

Don't do ref/mod emulation for KENTRY pages.
l2pte_minidata -> l2pte_minidata_p

(matt)

Revert previous - will fix userland with new sysctls for VM_MIN_ADDRESS/
VM_MAXADDESS.

(martin)

Add vm.min_address and vm.max_address which return VM_MIN_ADDRESS and
VM_MAXUSER_ADDRESS.

(matt)

powerpc64 support.

(christos)

powerpc64 support

(christos)

Make the .note section mergable and set proper "item" length.
XXX seems to be impossible to do that with gas w/o causing a (bogus)
warning - but the resulting object file is fine.

(martin)

Add missing EL_REFRESH

(christos)

Include param.h here - userland expects VM_* defines to work with just
this include, but PAGE_SIZE relies on PGSHIFT now, which is not defined
here.

Matt: please rearrange/cleanup differently if you prefer (but test build
src/tests/lib/libc/sys/t_mlock for example.)

(martin)

script to copy gcc to gcc.old, ready for importing.

(mrg)

prepare for moving GCC 4.5 into gcc.old:

- convert to using ${EXTERNAL_GCC_SUBDIR}
- define base-external-gpl3-gcc* subdir as GCC_SUBDIR
- use <bsd.init.mk> over <bsd.own.mk> for a bunch of places; mostly
  because it arranges for ../Makefile.inc to be included earlier, and
  don't bother including the latter if the former is already included.
- move all .PATH: settings after <bsd.{own,lib}.mk> so that all
  valid variables are set before it is evaluated
- rename mknative-gcc* to match their subdir name.

XXX the relationship between the Makefile.inc/Makefile.gcc_path files
    is kind of sketchy, it would be great if this was fixed.

(mrg)

Fix off by one in _bus_dma_{paddr_inrange,_to_paddr} when checking the
dma range.

(skrll)

Re-enable FSR_WRITE for armv6.

(skrll)

put the new EXTERNAL_GCC_SUBDIR under MKGCC.

(mrg)

Regen.

(ozaki-r)

Add Micrel PHY (KSZ9021RN)

The new driver micphy is almost same as ukphy except that
micphy has a fixup for cpsw; a PHY with cpsw has to adjust
RGMII signal timing.

Reviewed by christos@

(ozaki-r)

Make cpsw driver work without uboot support

On some eval boards such as BeagleBone, the cpsw device is initialized
rightly by the uboot of the boards so that the cpsw driver doesn't need
to do some initializations but works fine.

The patch adds initializations to make the driver work solely. It also
adds support for 1000BaseT (RGMII) PHY that is equipped on some boards,
e.g., CKB-3352.

Reviewed by christos@

(ozaki-r)

regen powerpc64

(christos)

powerpc64 needs USE_COMPILERCRTSTUFF=yes

(christos)

define EXTERNAL_GCC_SUBDIR based upon HAVE_GCC, for GCC >= 4.5,
and use it in the one place (bsd.lib.mk) it's used in share/mk.

(mrg)

Run constructor as part of network faction bootstrap, not device
(rumpdev no longer necessary)

(pooka)

Move pmap_recent_user to ci->ci_pmap_lastuser and
pmap_previous_active_lwp to ci->ci_lastlwp.  Fix some comments.

(matt)

Use MAP_ANON|MAP_PRIVATE for anonymous memory mapping,
e.g. Linux gets upset if just MAP_ANON.

(pooka)

Add more MMU registers

(matt)

hook up mmc card detect gpio

(jmcneill)

implement card detect

(jmcneill)

Move some static/global variables to here.

(matt)

Remove KERNEL_LOCK/KERNEL_UNLOCK

(matt)

Use PAGE_SIZE instead of 1000

(matt)

Add arm_cpu_max

(matt)

ci_cpu_id -> ci_arm_cpuid

(matt)

Add XN bit L2 large pages.

(matt)

Remove duplicate PMAP_PTE

(matt)

l2pte_valid -> l2pte_valid_p

(matt)

Replace 1000 with PAGE_SIZE

(matt)

Add support for PMAP_PTE to pmap_kenter_pa

(matt)

Use kmem instead of malloc.
Fix various corner cases with bounce buffers.
Use PMAP_NOCACHE instead of manipulating PTEs directly.

(matt)

Add PMAP_PTE flag.  (map page so it can be used as a PDE or PTE page)

(matt)

Use right name

(matt)

hook up MMC0 to status-led2

(jmcneill)

blink status LED during data transfers

(jmcneill)

Add awge
bump symtab space.

(matt)

Didn't want to comment out awin_dma_bootstrap

(matt)

Make this compile

(matt)

Deal with cubietruck GPIO differences.
Use its 2GB of RAM to distinguish it.

(matt)

Include latent support for coherent dma though for now it seems that isn't
possible.

(matt)

sdhc -> awinmmc

(jmcneill)

Fix bits/bytes mixup in rnd_getmore.

Remove some needless casts and fix format directives while here.

Bit/byte mixup noticed by pooka.

(riastradh)

If the in6 domain was not attached, do not attempt to process IPv6 packets.

(pooka)

Schedule only those netisr's that have registered handlers.

(pooka)

Only use FSR_WRITE on armv7 for now.

(matt)

Add support for saving the AVX-256 ymm registers during FPU context switches.
Add support for the forthcoming AVX-512 registers.
Code compiled with -mavx seems to work, but I've not tested context
  switches with live ymm registers.
There is a small cost on fork/exec (a larger area is copied/zerod),
  but I don't think the ymm registers are read/written unless they
  have been used.
The code use XSAVE on all cpus, I'm not brave enough to enable XSAVEOPT.

(dsl)

Add the XCR bits for snazzy upcoming features.
Define a mask for the fpu releated ones - only these wll be enabled.
The memory bound ones will need saving on every context switch.

(dsl)

Default to EXEC_ELF32

(matt)

emulate printflike properly to avoid fatal error on clang

(pooka)

add comments

(agc)

s/mcp48x1dac/mcp48x1dac(4)/

(rkujawa)

Note the new mcp48x1dac driver.

(rkujawa)

Add preliminary driver for Microchip MCP48x1 family of DACs.

Needs more testing and an appropriate API for such (non-audio) DACs. For now
one can set the output value using sysctl node. Also sysmon(4) will conveniently
report the voltage on Vout pin of the DAC. Note that this is not an actual
measured voltage, but merely calculated using equation from datasheet. Real
voltage might be different and depend on analog circuitry attached to the DAC.

(rkujawa)

Define ARM11_COMPAT_MMU (for now) to keep inline with  cpuconf.h changes

(skrll)

Fix thinko.

(matt)

Ensure that the top level sysctl nodes (kern, vfs, net, ...) exist before
the sysctl link sets are processed, and remove redundancy.

Shaves >13kB off of an amd64 GENERIC, not to mention >1k duplicate
lines of code.

(pooka)

Fix a 'stupido' that stopped (amongst other things) the cpu brand string
  being read.
The most obvious side effect the anita tests failed to detect they were
  running under qemu - so reported failures under qemu for things
  that qemu doesn't support.

(dsl)

Mark a potentially unused variable

(martin)

Remove an unused variable

(martin)

Mark a variable used only in diagnostics

(martin)

Mark a temporary as __USEd

(martin)

Build with FORTIFY_SOURCE

(roy)

Remove unused variable

(martin)

Fix obvious copy&pasto

(martin)

Note import of dhcpcd-6.3.0

(roy)

Sync

(roy)

Build new auth modules.

(roy)

Update generated files which were missed on xkeyboard-config 2.9 import.

http://mail-index.NetBSD.org/source-changes/2013/05/31/msg044255.html
http://mail-index.NetBSD.org/source-changes/2013/05/31/msg044256.html
http://mail-index.NetBSD.org/source-changes/2013/05/31/msg044258.html

XXX: should be regenerated during build but no perl and intltools in tools

(tsutsui)

Powerpc seems to expect minimum 8 byte alignment from malloc as well.

(martin)

ARMv6+ supplied a write bit in the FSR so use instead of reading the
instruction.

(matt)

Default ARM11 to ARM_MMU_V6N

(matt)

Add the ability to convert back to JSON in transit(1) - so, if desired,
JSON could be transported safely via the transit protocol.

Small fix for dictionary parsing.

Add more tests.

(agc)

Add missing include of <arm/cpufunc.h>.
Pointed out by Tero Koskinen.

(martin)

Add options COM_16750 to handle the IIR_BUSY interrupt.

(matt)

Add kern.{ostype,osrelease,osrevision,version} kern.domainname,
kern.rawpartition sysctl support to rump kernel.
Moved the sysctl support that is shared between rump and normal
kernels to init_sysctl_base.c as rump cannot use init_sysctl.c
in order to avoid code duplication. Agreed with pooka@.

(justin)

#include <arm/locore.h>

(matt)

enable MMC driver

(jmcneill)

Add Allwinner SD/MMC driver. No DMA yet.

(jmcneill)

Tag BE8 coredumps properly.

(matt)

fix typo

(matt)

If we are dumping a EABI coredump, set the version to EABI5.
If we are dumping a BE coredump on a BE8 system, set the BE8 flag.

(matt)

make more things static, remove dup decls

(christos)

make this compile

(christos)

Fix compat entries.

(joerg)

Update transit to version 20140223

+ properly escape valid JSON strings when converting

(agc)

+rump_wmd

(pooka)

Provide CAS variants for 16 and 8 bit when running with more that 1 cpu

(martin)

consistency in include protection

(christos)

Extract # of cores from L2CTRL for MULTIPROCESSOR kernels

(matt)

Add more inlines.  Default FSR ARMv6+ bits.  Add A7/A15 L2CTRL defs.

(matt)

Remove unneeded delays.  for ACHI port reg, act like ahcisatareg.h
(e.g. AHCI_P_AWIN_DMA(p))

(matt)

Provide cas_16 and cas_8 emulation via cas_32 and use that for mips64

(martin)

write to vendor specific dma reg relative to the port, not the base. sata works on cubieboard2 now

(jmcneill)

Remove the only use of __qdivrem inside kernels

(martin)